This application seeks to characterize critical steps in the proliferative signalling pathways activated by GM-CSF in related cytokines in primitive myeloid cells. At least some cases of myeloid leukemia likely result from over activity of such signal pathways. Identification of critical or unique steps in such cascades may allow development of pharmacologic interventions to specifically disrupt these pathways. The specific hypothesis of this application is that expression of the pim-1 gene project, a 33 kd serine/threonine kinase, is necessary for effective proliferative signalling through GM-CSF and related receptors. The applicant further hypothesizes that pim-1 expression serves as a specific indicator of activation of such pathways. The hypothesis will be investigated by constructing cell systems which allow for the over expression or blockade of pim-1 activity. Factor-dependent myeloid cell lines will be transfected with plasmid constructs or retroviral constructs to allow autonomous expression of the pim-1 gene product.These lines will then be examined for factor- independent growth, impaired differentiation, inhibition of apoptosis and homotypic aggregation. Similar cell lines will be created utilizing plasmids containing antisense pim-1 constructs to inhibit activity of the endogenous pim-1 gene. These cells will be studied for the responsiveness to GM-CSF and related cytokines and their propensity to differentiation.